CN108256429A - A kind of transmission tower object detection method using high spatial resolution single polarization SAR image - Google Patents

A kind of transmission tower object detection method using high spatial resolution single polarization SAR image Download PDF

Info

Publication number
CN108256429A
CN108256429A CN201711376811.1A CN201711376811A CN108256429A CN 108256429 A CN108256429 A CN 108256429A CN 201711376811 A CN201711376811 A CN 201711376811A CN 108256429 A CN108256429 A CN 108256429A
Authority
CN
China
Prior art keywords
image
pixel
sar image
single polarization
spatial resolution
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201711376811.1A
Other languages
Chinese (zh)
Inventor
马维青
武立平
龚浩
赵晓龙
武国亮
程远
吴强
冯智慧
牛彪
贾志义
方书博
陈汉超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan NARI Ltd
Yangquan Power Supply Co of State Grid Shanxi Electric Power Co Ltd
Original Assignee
Wuhan NARI Ltd
Yangquan Power Supply Co of State Grid Shanxi Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan NARI Ltd, Yangquan Power Supply Co of State Grid Shanxi Electric Power Co Ltd filed Critical Wuhan NARI Ltd
Priority to CN201711376811.1A priority Critical patent/CN108256429A/en
Publication of CN108256429A publication Critical patent/CN108256429A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of transmission tower object detection methods using high spatial resolution single polarization SAR image, include the following steps：According to original SAR image, corresponding radar power image is calculated；Multi-scale division is carried out according to the power image of SAR image, obtains the result of multi-scale division；With reference to multi-scale division as a result, the supervised classification method using object-oriented is classified, Preliminary detection result is obtained；According to the power image of SAR image, input picture is standardized using Gamma corrections, the image after being standardized；According to the image after standardization, HOG feature calculations are carried out；According to Preliminary detection result and the HOG features of image, target detection is carried out using SVM classifier.The advantage of the invention is that：It increases the dimension of feature, improves the precision of target detection, further reduced the false-alarm generated in target detection by calculating and the gradient orientation histogram of statistical picture regional area is come constitutive characteristic.

Description

A kind of transmission tower target detection using high spatial resolution single polarization SAR image Method
Method field
The present invention relates to transmission tower survey technology fields, and in particular to a kind of to be schemed using high spatial resolution single polarization SAR The transmission tower object detection method of picture.
Background method
Synthetic aperture radar is a kind of active imaging radar, it can be observed target to round-the-clock, have higher Resolution capability, remote sensing range is wide and with certain penetration capacity.With the continuous development of remote sensing technology, SAR technologies are wide It is applied to the multiple fields of military and civilian generally.In recent years, obtaining and being commonly used extensively with SAR image data, SAR image information extraction becomes the hot spot of research.Target detection is also obtained as the key technique in SAR image information extraction Vigorous growth is arrived.
However SAR image, due to being interfered by coherent speckle noise, atural object obscurity boundary, the interpretation of SAR image is relatively difficult, The coherent speckle noise in SAR image how to be inhibited to become the important prerequisite of SAR image target detection technique.Object--oriented method Effectively coherent spot can not only be inhibited to influence, also introduce more available features, help to understand the ground that image is included Object and target information, increase the precision of target detection, while reduce the false alarm rate in testing result.
Transmission tower is one of most important infrastructure in electric utility, and operating status decides the fortune of entire power grid Row stablizes and safety, carries out target detection to it and is of great significance.At present, in the inspection that transmission tower is carried out using remote sensing image In terms of surveying identification, domestic and international research work mainly has the method using the variation detection of multidate SAR image to carry out shape to power tower Become monitoring and carry out transmission line of electricity extraction using Mono temporal high-resolution full-polarization SAR, it is wider array of for practical ranges The single polarization SAR data of single phase, not yet propose it is a kind of utilize high spatial resolution single polarization SAR image carry out transmission tower The method of target detection.
Invention content
Present invention aim in view of the deficiencies of the prior art, provide a kind of utilization high spatial resolution single polarization The transmission tower object detection method of SAR image, by calculate and statistical picture regional area gradient orientation histogram come Constitutive characteristic increases the dimension of feature, improves the precision of target detection, further reduced the void generated in target detection It is alert.
To achieve the above object, a kind of transmission of electricity using high spatial resolution single polarization SAR image according to the present invention Shaft tower object detection method, includes the following steps：
Step 1：According to original SAR image, corresponding radar power image is calculated；
Step 2：Multi-scale division is carried out according to the power image of SAR image, obtains the result of multi-scale division；
Step 3：With reference to multi-scale division as a result, the supervised classification method using object-oriented is classified, at the beginning of obtaining Walk testing result；
Step 4：According to the power image of SAR image, input picture is standardized using Gamma corrections, is marked Image after standardization；
Step 5：According to the image after standardization, HOG feature calculations are carried out；
Step 6：According to Preliminary detection result and the HOG features of image, target detection is carried out using SVM classifier.
Further, in the step 1：
The calculation formula that i-th pixel calculates radar power image is：
In formula：RealbiRepresent the pixel value of single polarization SAR image real part, ImagbiRepresent single polarization SAR image imaginary part Pixel value, PiRepresent the corresponding power image pixel value of i-th of pixel.
Further, in the step 4：
Gamma correction the specific steps are：
Step 4.1：Power image is normalized, pixel value is converted into the real number between 0 and 1, is normalized Image later；
Step 4.2：Image after normalization is pre-compensated for, the image after being pre-compensated for；
Step 4.3：Renormalization is carried out to the image after precompensation, obtains Gamma correction results.
As preference, in the step 4.1：
I-th of pixel PiNormalized calculation formula is：
In formula：PmaxRepresent the maximum value of pixel in power image, PminRepresent the minimum value of pixel in power image, NiGeneration Pixel value after table normalization corresponding to i.
As preference, in the step 4.2：
The calculation formula to i-th pixel precompensation is：
In formula, NiFor the pixel value after normalization, Gamma is predetermined hyper parameter, YiIt is defeated after precompensation Go out value.
As preference, in the step 4.3：
The calculation formula of i-th of pixel renormalization is：
In formula：YmaxRepresent the maximum value of pixel in precompensation image, YminThe minimum value of pixel in precompensation image is represented, P′iRepresent the pixel value corresponding to i after renormalization.
Further, in the step 5：
It is described calculate HOG features the specific steps are：
Step 5.1：The gradient of each pixel of image after Gamma is corrected is calculated, including size and Orientation；
Step 5.2：Small Cell is divided an image into, counts the histogram of gradients of each Cell；
Step 5.3：A Block will be formed per several Cell, and echelon's histogram of Block will be normalized, it will The histogram of gradients of all Block, which is together in series, obtains the HOG features of image.
As preference, in the step 5.1：
The gradient for calculating each pixel of image after Gamma is corrected, specific calculation formula are：
For pixel (x, y) in image：
Gx(x, y)=H (x+1, y)-H (x-1, y)
Gy(x, y)=H (x, y+1)-H (x, y-1)
In formula:Gx(x, y), Gy(x, y), H (x, y) represent the horizontal direction ladder at pixel (x, y) in input picture respectively Degree, vertical gradient and pixel value.Gradient magnitude and gradient direction at pixel (x, y) are respectively：
As preference, in the step 5.2：
Divide an image into small Cell, count each Cell histogram of gradients the specific steps are：By image according to one Fixed window size is divided into Cell, and the gradient magnitude in each Cell is weighted throwing according to place gradient direction section Shadow obtains the histogram of gradients of each Cell.
As preference, in the step 5.3：
A Block will be formed per several Cell, and to specific steps that echelon's histogram of Block is normalized For：According to certain step-length, several Cell are formed into a Block；All histogram of gradients in same Block are carried out Normalized；The normalization of all Block is completed to get to the histogram of gradients of entire image.
The advantage of the invention is that：It uses multi-scale division as the method with object-oriented supervised classification as preliminary Detection means, not only effectively remains the original boundaries of ground object target, but also the statistical information being utilized in single polarization SAR image, While inhibiting coherent speckle noise, the accuracy of single polarization SAR image transmission tower target detection is improved；Use HOG features Preliminary detection result is detected again, spy is formed with the gradient orientation histogram of statistical picture regional area by calculating Sign, increases the dimension of feature, improves the precision of target detection, further reduced the false-alarm generated in target detection.
Description of the drawings
Fig. 1 is the work flow diagram of the present invention；
Fig. 2 is the L-band VV polarization mode power diagrams that airborne UAVSAR is obtained；
Fig. 3 is the result figure that multi-scale division obtains；
Fig. 4 is the Preliminary detection result figure obtained with the supervised classification of segmentation result and object-oriented；
Fig. 5 is the transmission tower object detection results figure of airborne L-band VV Polarimetric SAR Images.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments：
Such as Fig. 1, a kind of transmission tower object detection method using high spatial resolution single polarization SAR image, including Following steps：
Step 1：According to original SAR image, corresponding radar power image is calculated：
The calculation formula that i-th pixel calculates radar power image is：
In formula：RealbiRepresent the pixel value of single polarization SAR image real part, ImagbiRepresent single polarization SAR image imaginary part Pixel value, PiRepresent the corresponding power image pixel value of i-th of pixel.
Step 2：Multi-scale division is carried out according to the power image of SAR image, obtains the result of multi-scale division；
Step 3：With reference to multi-scale division as a result, the supervised classification method using object-oriented is classified, at the beginning of obtaining Walk testing result；
Step 4：According to the power image of SAR image, input picture is standardized using Gamma corrections, is marked Image after standardization：
Gamma correction the specific steps are：
Step 4.1：Power image is normalized, pixel value is converted into the real number between 0 and 1, is normalized Image later：
I-th of pixel PiNormalized calculation formula is：
In formula：PmaxRepresent the maximum value of pixel in power image, PminRepresent the minimum value of pixel in power image, NiGeneration Pixel value after table normalization corresponding to i.
Step 4.2：Image after normalization is pre-compensated for, the image after being pre-compensated for：
The calculation formula to i-th pixel precompensation is：
In formula, NiFor the pixel value after normalization, Gamma is predetermined hyper parameter, YiIt is defeated after precompensation Go out value.
Step 4.3：Renormalization is carried out to the image after precompensation, obtains Gamma correction results：
The calculation formula of i-th of pixel renormalization is：
In formula：YmaxRepresent the maximum value of pixel in precompensation image, YminThe minimum value of pixel in precompensation image is represented, P′iRepresent the pixel value corresponding to i after renormalization.
Step 5：According to the image after standardization, HOG feature calculations are carried out：
It is described calculate HOG features the specific steps are：
Step 5.1：The gradient of each pixel of image after Gamma is corrected is calculated, including size and Orientation：
The gradient for calculating each pixel of image after Gamma is corrected, specific calculation formula are：
For pixel (x, y) in image：
Gx(x, y)=H (x+1, y)-H (x-1, y)
Gy(x, y)=H (x, y+1)-H (x, y-1)
In formula:Gx(x, y), Gy(x, y), H (x, y) represent the horizontal direction ladder at pixel (x, y) in input picture respectively Degree, vertical gradient and pixel value.Gradient magnitude and gradient direction at pixel (x, y) are respectively：
Step 5.2：Small Cell is divided an image into, counts the histogram of gradients of each Cell：
Divide an image into small Cell, count each Cell histogram of gradients the specific steps are：By image according to one Fixed window size is divided into Cell, and the gradient magnitude in each Cell is weighted throwing according to place gradient direction section Shadow obtains the histogram of gradients of each Cell.
Step 5.3：A Block will be formed per several Cell, and echelon's histogram of Block will be normalized, it will The histogram of gradients of all Block, which is together in series, obtains the HOG features of image：
A Block will be formed per several Cell, and to specific steps that echelon's histogram of Block is normalized For：According to certain step-length, several Cell are formed into a Block；All histogram of gradients in same Block are carried out Normalized；The normalization of all Block is completed to get to the histogram of gradients of entire image.
Step 6：According to Preliminary detection result and the HOG features of image, target detection is carried out using SVM classifier.
The present invention is in actual use：
1. example content：
The result of present example experiment is as shown in Figure 2-5.Fig. 2 is the L-band VV polarization modes that airborne UAVSAR is obtained Power image, distance are each about 8m to azimuth resolution.Fig. 3 is the result that multi-scale division obtains；Fig. 4 is to divide knot The Preliminary detection result that the supervised classification of fruit and object-oriented obtains；Fig. 5 is the testing result that the present invention obtains.
2. experimental result and analysis：
It can be seen that from Fig. 2, Fig. 3, Fig. 4 using multi-scale division and Object--oriented method, greatly SAR inhibited to scheme Coherent speckle noise as in, is more utilized the statistical information in SAR image；And the means of multistage detection are used, it is improving While verification and measurement ratio, the false-alarm in testing result is further reduced.
The method that the present invention uses multi-scale division, preferably remains the boundary of target, utilizes the preliminary of object-oriented Detection successfully inhibits the coherent speckle noise in SAR image, reduces the difficulty of target detection, and multistage is used after extracting HOG features The detection mode of multiple features achievees the purpose that reduce false-alarm while transmission tower target detection accuracy is improved.
Finally it is pointed out that above example is only the more representational example of the present invention.It is clear that the invention is not restricted to Above-described embodiment, acceptable there are many deform.All methods according to the present invention are substantially made to the above embodiment any simple Modification, equivalent variations and modification, are considered as belonging to the scope of protection of the present invention.

Claims (10)

1. a kind of transmission tower object detection method using high spatial resolution single polarization SAR image, it is characterised in that：Including Following steps：
Step 1：According to original SAR image, corresponding radar power image is calculated；
Step 2：Multi-scale division is carried out according to the power image of SAR image, obtains the result of multi-scale division；
Step 3：With reference to multi-scale division as a result, the supervised classification method using object-oriented is classified, tentatively examined Survey result；
Step 4：According to the power image of SAR image, input picture is standardized using Gamma corrections, is standardized Image afterwards；
Step 5：According to the image after standardization, HOG feature calculations are carried out；
Step 6：According to Preliminary detection result and the HOG features of image, target detection is carried out using SVM classifier.
2. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 1 Method, it is characterised in that：In the step 1：
The calculation formula that i-th pixel calculates radar power image is：
In formula：RealbiRepresent the pixel value of single polarization SAR image real part, ImagbiRepresent the pixel of single polarization SAR image imaginary part Value, PiRepresent the corresponding power image pixel value of i-th of pixel.
3. a kind of transmission tower target using high spatial resolution single polarization SAR image according to claim 1 or 2 is examined Survey method, it is characterised in that：In the step 4：
Gamma correction the specific steps are：
Step 4.1：Power image is normalized, pixel value is converted into the real number between 0 and 1, after obtaining normalization Image；
Step 4.2：Image after normalization is pre-compensated for, the image after being pre-compensated for；
Step 4.3：Renormalization is carried out to the image after precompensation, obtains Gamma correction results.
4. it is examined according to a kind of transmission tower target using high spatial resolution single polarization SAR image described in claim 3 Survey method, it is characterised in that：In the step 4.1：
I-th of pixel PiNormalized calculation formula is：
In formula：PmaxRepresent the maximum value of pixel in power image, PminRepresent the minimum value of pixel in power image, NiRepresentative is returned Pixel value after one change corresponding to i.
5. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 4 Method, it is characterised in that：In the step 4.2：
The calculation formula to i-th pixel precompensation is：
In formula, NiFor the pixel value after normalization, Gamma is predetermined hyper parameter, YiFor the output after precompensation Value.
6. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 5 Method, it is characterised in that：In the step 4.3：
The calculation formula of i-th of pixel renormalization is：
In formula：YmaxRepresent the maximum value of pixel in precompensation image, YminRepresent the minimum value of pixel in precompensation image, P 'iGeneration Pixel value after table renormalization corresponding to i.
7. a kind of transmission of electricity using high spatial resolution single polarization SAR image according to any one in claim 4~6 Shaft tower object detection method, it is characterised in that：In the step 5：
It is described calculate HOG features the specific steps are：
Step 5.1：The gradient of each pixel of image after Gamma is corrected is calculated, including size and Orientation；
Step 5.2：Small Cell is divided an image into, counts the histogram of gradients of each Cell；
Step 5.3：A Block will be formed per several Cell, and echelon's histogram of Block will be normalized, it will be all The histogram of gradients of Block, which is together in series, obtains the HOG features of image.
8. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 7 Method, it is characterised in that：In the step 5.1：
The gradient for calculating each pixel of image after Gamma is corrected, specific calculation formula are：
For pixel (x, y) in image：
Gx(x, y)=H (x+1, y)-H (x-1, y)
Gy(x, y)=H (x, y+1)-H (x, y-1)
In formula：Gx(x, y), Gy(x, y), H (x, y) represent the horizontal direction gradient in input picture at pixel (x, y) respectively, Vertical gradient and pixel value.Gradient magnitude and gradient direction at pixel (x, y) are respectively：
9. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 8 Method, it is characterised in that：In the step 5.2：
Divide an image into small Cell, count each Cell histogram of gradients the specific steps are：By image according to certain Window size is divided into Cell, and the gradient magnitude in each Cell is weighted projection according to place gradient direction section, is obtained To the histogram of gradients of each Cell.
10. a kind of transmission tower target detection using high spatial resolution single polarization SAR image according to claim 9 Method, it is characterised in that：In the step 5.3：
Will per several Cell form a Block, and to echelon's histogram of Block be normalized the specific steps are：It presses According to certain step-length, several Cell are formed into a Block；Normalizing is carried out to all histogram of gradients in same Block Change is handled；The normalization of all Block is completed to get to the histogram of gradients of entire image.
CN201711376811.1A 2017-12-19 2017-12-19 A kind of transmission tower object detection method using high spatial resolution single polarization SAR image Pending CN108256429A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711376811.1A CN108256429A (en) 2017-12-19 2017-12-19 A kind of transmission tower object detection method using high spatial resolution single polarization SAR image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711376811.1A CN108256429A (en) 2017-12-19 2017-12-19 A kind of transmission tower object detection method using high spatial resolution single polarization SAR image

Publications (1)

Publication Number Publication Date
CN108256429A true CN108256429A (en) 2018-07-06

Family

ID=62723667

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711376811.1A Pending CN108256429A (en) 2017-12-19 2017-12-19 A kind of transmission tower object detection method using high spatial resolution single polarization SAR image

Country Status (1)

Country Link
CN (1) CN108256429A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109254287A (en) * 2018-10-08 2019-01-22 国网经济技术研究院有限公司 Method for detecting power transmission conductor based on synthetic aperture radar image
CN109359509A (en) * 2018-08-27 2019-02-19 西安电子科技大学 Based on the parallel polarization SAR terrain classification method of OpenCL
CN109376623A (en) * 2018-10-08 2019-02-22 国网经济技术研究院有限公司 Method for detecting transmission tower based on synthetic aperture radar image
CN110708511A (en) * 2019-10-17 2020-01-17 山东浪潮人工智能研究院有限公司 Monitoring video compression method based on image target detection
CN113313162A (en) * 2021-05-25 2021-08-27 国网河南省电力公司电力科学研究院 Method and system for detecting multi-scale feature fusion target

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104036530A (en) * 2014-05-21 2014-09-10 中国科学院电子学研究所 Processing method and device for SAR images
CN106407900A (en) * 2016-08-31 2017-02-15 上海交通大学 Multi-source aerial photo-based abnormal scene identification method
CN106651880A (en) * 2016-12-27 2017-05-10 首都师范大学 Method for detecting marine moving target of thermal infrared remote sensing image based on multi-feature fusion
CN106709463A (en) * 2016-12-29 2017-05-24 武汉大学 Hierarchical semantics-based SAR image classification method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104036530A (en) * 2014-05-21 2014-09-10 中国科学院电子学研究所 Processing method and device for SAR images
CN106407900A (en) * 2016-08-31 2017-02-15 上海交通大学 Multi-source aerial photo-based abnormal scene identification method
CN106651880A (en) * 2016-12-27 2017-05-10 首都师范大学 Method for detecting marine moving target of thermal infrared remote sensing image based on multi-feature fusion
CN106709463A (en) * 2016-12-29 2017-05-24 武汉大学 Hierarchical semantics-based SAR image classification method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
刘奕 等: "基于高分辨率ＳＡＲ影像的高压输电线路杆塔检测方法", 《电力科学与技术学报》 *
蒋丹丹 等: "基于单极化TerraSAR_X影像提取建筑区研究", 《地理与地理信息科学》 *
薛笑荣: "《SAR图像处理技术研究》", 31 August 2017, 科学技术文献出版社 *
邓磊: "《SAR图像处理方法》", 28 February 2009, 测绘出版社 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109359509A (en) * 2018-08-27 2019-02-19 西安电子科技大学 Based on the parallel polarization SAR terrain classification method of OpenCL
CN109359509B (en) * 2018-08-27 2021-09-28 西安电子科技大学 OpenCL parallel-based polarized SAR terrain classification method
CN109254287A (en) * 2018-10-08 2019-01-22 国网经济技术研究院有限公司 Method for detecting power transmission conductor based on synthetic aperture radar image
CN109376623A (en) * 2018-10-08 2019-02-22 国网经济技术研究院有限公司 Method for detecting transmission tower based on synthetic aperture radar image
CN109376623B (en) * 2018-10-08 2021-12-21 国网经济技术研究院有限公司 Method for detecting transmission tower based on synthetic aperture radar image
CN110708511A (en) * 2019-10-17 2020-01-17 山东浪潮人工智能研究院有限公司 Monitoring video compression method based on image target detection
CN113313162A (en) * 2021-05-25 2021-08-27 国网河南省电力公司电力科学研究院 Method and system for detecting multi-scale feature fusion target

Similar Documents

Publication Publication Date Title
CN108256429A (en) A kind of transmission tower object detection method using high spatial resolution single polarization SAR image
Liu et al. Automatic cloud detection for all-sky images using superpixel segmentation
CN108416791A (en) A kind of monitoring of parallel institution moving platform pose and tracking based on binocular vision
CN100507603C (en) Hyperspectral image abnormal point detection method based on selective kernel principal component analysis
Wang et al. Recognition and drop-off detection of insulator based on aerial image
CN106529416A (en) Electric-power line detection method and system based on millimeter wave radar decision tree classification
CN110084165A (en) The intelligent recognition and method for early warning of anomalous event under the open scene of power domain based on edge calculations
CN106022345B (en) A kind of high voltage isolator state identification method based on Hough forest
CN103177241B (en) The method utilizing video image processing technology location power transmission line spacer
CN104700110A (en) Plant covering information extracting method based on perfect polarization SAR images
CN105005767A (en) Microwave remote sensing image based forest type identification method
CN108052929A (en) Parking space state detection method, system, readable storage medium storing program for executing and computer equipment
CN107273799A (en) A kind of indoor orientation method and alignment system
CN104978743A (en) Multi-core parallel SAR (synthetic aperture radar) image change information real-time extraction method and extraction device
CN105787470A (en) Method for detecting power transmission line tower in image based on polymerization multichannel characteristic
CN107390259A (en) A kind of nuclide identification method based on SVD and SVM
CN103714319A (en) Video intrusion detection method used for power line external force damage-prevention system
CN115620239A (en) Point cloud and video combined power transmission line online monitoring method and system
CN104156981B (en) A kind of automatic real-time detection method of solar radio burst based on image procossing
Zeng et al. Power transmission tower detection based on polar coordinate semivariogram in high-resolution SAR image
CN104166975A (en) Low-altitude infrared target detection algorithm in complex environment
CN109271864A (en) A kind of crowd density estimation method based on wavelet transformation and support vector machines
CN106228192B (en) A kind of transmission line forest fire domain identification method analyzed based on double-threshold cluster
CN106251351B (en) A kind of transmission line forest fire monitoring threshold computational methods based on Cloud transform
CN105551017A (en) Transmission line forest fire target extraction method on the basis of spatio-temporal union

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20180706